Sunday 28 January 2007

Error Analysis Saves Lives

So, here it is. Another night-before-lab-book-gets-handed-in, and I am scrawling away, overlooking the criminal nature of my layout, accuracy, maths, logical flaws and so forth (but never my English), in a half-hearted attempt not to fail as badly as last time. It's a small mercy that this particular experiment hasn't had too much in the way of errors, however what it has had has been as excruciatingly painful as usual.

When I run into a tight corner with my write-up, as I nearly always do, I plead to my University Godparents for guidance - Wikipedia and Google. They have been of little help to me tonight, however through them I have apparently learned the importance of error analysis. Columbia University, New York assures me that "bad things can happen if error analysis is ignored". To back up their claim, which in this context is shamefully less than scientific, they have provided the following picture of the derailment at Gare Montparnasse, Paris, 1895:



There you have it. I must press on to avoid causing any such disasters. I'm a little disappointed I was not warned about this in my lab script.

Links of the day: Columbia University's Error Analysis Tutorial

No comments: